I fixed the code from the previous week's audio analysis to look at a much shorter audio sample, and the results for the FFT and the Hanning window make a lot more sense now. I also added a Gaussian convolution to the original signal and also to of the frequency domain graph (out of curiosity), but I'm not sure it is functioning correctly in either.
In the other file, I apply a 2D Gaussian kernel to an image to blur it, but made use of the convolution theorem to perform the convolution by FFT.
![[paper clip]](/cours/static/images/paper_clip_tilt.png)
| last modified | size | ||
 |
AudioAnalysis2.html | Tue Jul 28 2026 10:11 am | 359K |
 |
AudioAnalysis2.ipynb | Tue Jul 28 2026 10:11 am | 100K |
|
GaussianBlurWith_FFT.html | Tue Jul 28 2026 10:11 am | 552K |
|
GaussianBlurWith_FFT.ipynb | Tue Jul 28 2026 10:11 am | 302K |
|
jim_2d.html | Tue Jul 28 2026 10:11 am | 464K |
|
jim_2d.ipynb | Tue Jul 28 2026 10:11 am | 216K |
